Skip to content

Instantly share code, notes, and snippets.

@tej87681088

tej87681088/TEJAPI_Python_Fisher_Strategy.ipynb

Created October 24, 2022 01:31
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save tej87681088/13a00a55c739d4b12a234e21812b7589 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save tej87681088/13a00a55c739d4b12a234e21812b7589 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
TEJAPI_Python_Fisher_Strategy.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import tejapi as tej\n",
"import pandas as pd\n",
"import numpy as np\n",
"import datetime\n",
"import matplotlib.pyplot as plt\n",
"\n",
"tej.ApiConfig.api_key = 'Your Key'\n",
"\n",
"plt.rcParams['font.sans-serif'] = ['Taipei Sans TC Beta']"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 找到所需的會計科目名稱：\n",
"### R505負債比率、單月營收、R108稅後淨利率、PSR:市值/近一年月營收、PRR:市值/近一年3368研發費用"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"financial_name = tej.get('TWN/AIACC', chinese_column_name=True, paginate=True) #找尋所需的會計科目\n",
"\n",
"financial_dict = ['負債比率', '稅後淨利率', '研發費用']\n",
"\n",
"a = []\n",
"for i in financial_dict:\n",
" a.append(financial_name[financial_name['中文全稱'].str.contains(i)]['會計科目'].mode().to_list())"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[['R505'], ['R108'], ['3368']]"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"a "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 下載12~22年間所有上市櫃相關財報資料"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"#上市櫃公司代號\n",
"code = tej.get('TWN/ANPRCSTD', mdate={'lt':'2022-10-14'}, chinese_column_name=True, paginate=True)\n",
"all_code = code[(code['證券種類名稱'] == '普通股') & (code['上市別'].isin(['TSE', 'OTC']))]['證券碼'].to_list()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"#下載近十年的財務數據\n",
"financial_data = tej.get('TWN/AIFINQ', \n",
" coid=all_code, \n",
" mdate={'gt': '2012-01-01', 'lt':'2022-10-14'}, \n",
" acc_code=a,\n",
" chinese_column_name=True, \n",
" paginate=True)\n",
"financial_data = financial_data.reset_index(drop=True)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\terry\\AppData\\Local\\Temp\\ipykernel_7908\\4154287080.py:10: FutureWarning: Using .astype to convert from timezone-aware dtype to timezone-naive dtype is deprecated and will raise in a future version. Use obj.tz_localize(None) or obj.tz_convert('UTC').tz_localize(None) instead\n",
" revenue_data = revenue_data.astype({'年/月':'datetime64[ns]'})\n"
]
}
],
"source": [
"#下載近十年的月營收數據\n",
"revenue_data = tej.get('TWN/ASALE', \n",
" coid=all_code, \n",
" mdate={'gt': '2011-12-31', 'lt':'2022-10-14'}, \n",
" opts={'columns': ['coid', 'mdate', 'd0001']},\n",
" chinese_column_name=True, \n",
" paginate=True)\n",
"revenue_data.rename(columns={'年月': '年/月'}, inplace=True)\n",
"revenue_data['公司'] = revenue_data['公司'].astype(int)\n",
"revenue_data = revenue_data.astype({'年/月':'datetime64[ns]'})"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 下載12~22年間所有股價資料"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\terry\\AppData\\Local\\Temp\\ipykernel_7908\\24625519.py:16: FutureWarning: Using .astype to convert from timezone-aware dtype to timezone-naive dtype is deprecated and will raise in a future version. Use obj.tz_localize(None) or obj.tz_convert('UTC').tz_localize(None) instead\n",
" price = price.astype({'年/月':'datetime64[ns]'})\n"
]
}
],
"source": [
"m = pd.date_range('2011-12-31', '2022-11-01', freq='1M', inclusive='both').to_list()\n",
"\n",
"price = pd.DataFrame()\n",
"\n",
"for i in range(1, len(m)):\n",
" price = pd.concat([price, tej.get('TWN/APRCD1', \n",
" coid=all_code, \n",
" mdate={'gt': m[i-1], 'lt':m[i]}, \n",
" opts={'columns': ['coid', 'mdate', 'close_adj', 'mv']},\n",
" chinese_column_name=True, \n",
" paginate=True)])\n",
"\n",
"price = price.reset_index(drop=True)\n",
"price = price.rename(columns={'證券代碼':'公司', '年月日':'年/月'})\n",
"price['公司'] = price['公司'].astype(int)\n",
"price = price.astype({'年/月':'datetime64[ns]'})"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>公司</th>\n",
" <th>年/月</th>\n",
" <th>收盤價(元)</th>\n",
" <th>市值(百萬元)</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1101</td>\n",
" <td>2012-01-02</td>\n",
" <td>14.5382</td>\n",
" <td>128119</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1101</td>\n",
" <td>2012-01-03</td>\n",
" <td>14.8315</td>\n",
" <td>130703</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1101</td>\n",
" <td>2012-01-04</td>\n",
" <td>14.9153</td>\n",
" <td>131441</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1101</td>\n",
" <td>2012-01-05</td>\n",
" <td>14.9153</td>\n",
" <td>131441</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1101</td>\n",
" <td>2012-01-06</td>\n",
" <td>14.6429</td>\n",
" <td>129042</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3776287</th>\n",
" <td>9962</td>\n",
" <td>2022-10-17</td>\n",
" <td>13.6500</td>\n",
" <td>1232</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3776288</th>\n",
" <td>9962</td>\n",
" <td>2022-10-18</td>\n",
" <td>13.8000</td>\n",
" <td>1245</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3776289</th>\n",
" <td>9962</td>\n",
" <td>2022-10-19</td>\n",
" <td>14.1000</td>\n",
" <td>1272</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3776290</th>\n",
" <td>9962</td>\n",
" <td>2022-10-20</td>\n",
" <td>14.3000</td>\n",
" <td>1290</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3776291</th>\n",
" <td>9962</td>\n",
" <td>2022-10-21</td>\n",
" <td>14.3500</td>\n",
" <td>1295</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>3776292 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" 公司 年/月 收盤價(元) 市值(百萬元)\n",
"0 1101 2012-01-02 14.5382 128119\n",
"1 1101 2012-01-03 14.8315 130703\n",
"2 1101 2012-01-04 14.9153 131441\n",
"3 1101 2012-01-05 14.9153 131441\n",
"4 1101 2012-01-06 14.6429 129042\n",
"... ... ... ... ...\n",
"3776287 9962 2022-10-17 13.6500 1232\n",
"3776288 9962 2022-10-18 13.8000 1245\n",
"3776289 9962 2022-10-19 14.1000 1272\n",
"3776290 9962 2022-10-20 14.3000 1290\n",
"3776291 9962 2022-10-21 14.3500 1295\n",
"\n",
"[3776292 rows x 4 columns]"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"price "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 指標計算: 負債比率<35%、最近5年平均營收成長率≧15%、最近5年平均稅後淨利率≧5%、股價營收比(PSR)≦1.5、股價／研發費用比(PRR) ≦15"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\terry\\AppData\\Local\\Temp\\ipykernel_7908\\1026380122.py:10: FutureWarning: Using .astype to convert from timezone-aware dtype to timezone-naive dtype is deprecated and will raise in a future version. Use obj.tz_localize(None) or obj.tz_convert('UTC').tz_localize(None) instead\n",
" financial_data1 = financial_data1.astype({'年/月':'datetime64[ns]'})\n"
]
}
],
"source": [
"financial_data1 = financial_data.copy()\n",
"financial_data1['數值'] = financial_data1['數值'].astype(float)\n",
"\n",
"financial_data1['會計科目'] = financial_data1['會計科目'].map({'R505':'負債比率', #會計科目改名\n",
" 'R108':'稅後淨利率',\n",
" '3368':'研發費用'})\n",
" \n",
"financial_data1 = financial_data1.pivot_table(index=['公司', '年/月'], columns='會計科目', values='數值').reset_index() #格式轉置\n",
"financial_data1['公司'] = financial_data1['公司'].astype(int)\n",
"financial_data1 = financial_data1.astype({'年/月':'datetime64[ns]'})\n",
"financial_data1 = financial_data1.fillna(0)"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"financial_data1 = pd.merge_asof(revenue_data.sort_values('年/月'), financial_data1.sort_values('年/月'), on='年/月', by='公司', direction='backward').sort_values(['公司', '年/月'])\n",
"financial_data1 = pd.merge_asof(financial_data1.sort_values('年/月'), price.sort_values('年/月'), on='年/月', by='公司', direction='nearest').sort_values(['公司', '年/月'])"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [],
"source": [
"financial_data1['營收年成長率'] = financial_data1['單月營收(千元)'] / financial_data1['單月營收(千元)'].shift(12)\n",
"\n",
"financial_data1['平均5年營收成長率'] = financial_data1.groupby('公司')['營收年成長率'].transform(lambda x: x.rolling(60).mean())\n",
"financial_data1['平均5年稅後淨利率'] = financial_data1.groupby('公司')['稅後淨利率'].transform(lambda x: x.rolling(60).mean())\n",
"\n",
"financial_data1['近一年營收總合'] = financial_data1.groupby('公司')['單月營收(千元)'].transform(lambda x: x.rolling(12).sum()/1000)\n",
"financial_data1['股價營收比'] = financial_data1['市值(百萬元)'] / financial_data1['近一年營收總合']\n",
"\n",
"financial_data1['近一年研發總合'] = financial_data1.groupby('公司')['研發費用'].transform(lambda x: x.rolling(12).sum()/1000)\n",
"financial_data1['股價研發比'] = financial_data1['市值(百萬元)'] / financial_data1['近一年研發總合']"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [],
"source": [
"financial_data2 = financial_data1[['公司', '年/月', '負債比率', '收盤價(元)', '平均5年營收成長率', '平均5年稅後淨利率', '股價營收比', '股價研發比']].reset_index(drop=True)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>公司</th>\n",
" <th>年/月</th>\n",
" <th>負債比率</th>\n",
" <th>收盤價(元)</th>\n",
" <th>平均5年營收成長率</th>\n",
" <th>平均5年稅後淨利率</th>\n",
" <th>股價營收比</th>\n",
" <th>股價研發比</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1101</td>\n",
" <td>2012-01-01</td>\n",
" <td>NaN</td>\n",
" <td>14.5382</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1101</td>\n",
" <td>2012-02-01</td>\n",
" <td>NaN</td>\n",
" <td>15.2085</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1101</td>\n",
" <td>2012-03-01</td>\n",
" <td>47.80</td>\n",
" <td>15.6066</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1101</td>\n",
" <td>2012-04-01</td>\n",
" <td>47.80</td>\n",
" <td>14.3287</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1101</td>\n",
" <td>2012-05-01</td>\n",
" <td>47.80</td>\n",
" <td>14.8943</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>204457</th>\n",
" <td>9962</td>\n",
" <td>2022-05-01</td>\n",
" <td>24.53</td>\n",
" <td>17.2820</td>\n",
" <td>1.121627</td>\n",
" <td>2.163000</td>\n",
" <td>0.542494</td>\n",
" <td>inf</td>\n",
" </tr>\n",
" <tr>\n",
" <th>204458</th>\n",
" <td>9962</td>\n",
" <td>2022-06-01</td>\n",
" <td>31.41</td>\n",
" <td>15.6690</td>\n",
" <td>1.128230</td>\n",
" <td>2.312833</td>\n",
" <td>0.480278</td>\n",
" <td>inf</td>\n",
" </tr>\n",
" <tr>\n",
" <th>204459</th>\n",
" <td>9962</td>\n",
" <td>2022-07-01</td>\n",
" <td>31.41</td>\n",
" <td>12.2000</td>\n",
" <td>1.134070</td>\n",
" <td>2.462667</td>\n",
" <td>0.338844</td>\n",
" <td>inf</td>\n",
" </tr>\n",
" <tr>\n",
" <th>204460</th>\n",
" <td>9962</td>\n",
" <td>2022-08-01</td>\n",
" <td>31.41</td>\n",
" <td>15.8500</td>\n",
" <td>1.149276</td>\n",
" <td>2.612500</td>\n",
" <td>0.422695</td>\n",
" <td>inf</td>\n",
" </tr>\n",
" <tr>\n",
" <th>204461</th>\n",
" <td>9962</td>\n",
" <td>2022-09-01</td>\n",
" <td>31.41</td>\n",
" <td>14.8500</td>\n",
" <td>1.151498</td>\n",
" <td>2.711833</td>\n",
" <td>0.395983</td>\n",
" <td>inf</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>204462 rows × 8 columns</p>\n",
"</div>"
],
"text/plain": [
" 公司 年/月 負債比率 收盤價(元) 平均5年營收成長率 平均5年稅後淨利率 股價營收比 股價研發比\n",
"0 1101 2012-01-01 NaN 14.5382 NaN NaN NaN NaN\n",
"1 1101 2012-02-01 NaN 15.2085 NaN NaN NaN NaN\n",
"2 1101 2012-03-01 47.80 15.6066 NaN NaN NaN NaN\n",
"3 1101 2012-04-01 47.80 14.3287 NaN NaN NaN NaN\n",
"4 1101 2012-05-01 47.80 14.8943 NaN NaN NaN NaN\n",
"... ... ... ... ... ... ... ... ...\n",
"204457 9962 2022-05-01 24.53 17.2820 1.121627 2.163000 0.542494 inf\n",
"204458 9962 2022-06-01 31.41 15.6690 1.128230 2.312833 0.480278 inf\n",
"204459 9962 2022-07-01 31.41 12.2000 1.134070 2.462667 0.338844 inf\n",
"204460 9962 2022-08-01 31.41 15.8500 1.149276 2.612500 0.422695 inf\n",
"204461 9962 2022-09-01 31.41 14.8500 1.151498 2.711833 0.395983 inf\n",
"\n",
"[204462 rows x 8 columns]"
]
},
"execution_count": 48,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"financial_data2"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 策略篩選"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [],
"source": [
"def condition(df, n1, n2, n3, n4, n5):\n",
" price_T = price.pivot_table(index='年/月', columns='公司', values='收盤價(元)')\n",
"\n",
" target_stock = df[(df['負債比率'] < n1) & (df['平均5年營收成長率'] >= n2) & (df['平均5年稅後淨利率'] >= n3) & (df['股價營收比'] <= n4) & (df['股價研發比'] <= n5)]\n",
"\n",
" target_Q = list(target_stock.groupby('年/月')[['公司', '年/月', '收盤價(元)']])\n",
"\n",
" ret, daily_ret = [], []\n",
" for i in range(len(target_Q)):\n",
" daily_ret.append(price_T.loc[target_Q[i][1].iloc[0]['年/月']:target_Q[i][1].iloc[0]['年/月']+pd.DateOffset(months=1), target_Q[i][1]['公司'].to_list()])\n",
" ret.append(price_T.loc[target_Q[i][1].iloc[0]['年/月']:target_Q[i][1].iloc[0]['年/月']+pd.DateOffset(months=1), target_Q[i][1]['公司'].to_list()].pct_change().mean(axis=1))\n",
"\n",
" ret_table = pd.concat(ret)\n",
"\n",
" ret_table = pd.DataFrame(columns=['投組日報酬'], data=ret_table).dropna()\n",
"\n",
" ret_table['投組累積報酬'] = (ret_table['投組日報酬'] +1).cumprod()\n",
"\n",
"\n",
"\n",
" ret_table.reset_index(inplace=True)\n",
"\n",
" print('年化報酬率:',\"%6.3f\" % (ret_table['投組日報酬'].mean()*252))\n",
" print('年化標準差:',\"%6.3f\" % (ret_table['投組日報酬'].std()*np.sqrt(252)))\n",
" print('年化夏普:',\"%6.4f\" % (ret_table['投組日報酬'].mean()*252 / (ret_table['投組日報酬'].std()*np.sqrt(252))))\n",
"\n",
"\n",
" return ret_table, target_stock[['公司', '年/月', '收盤價(元)']], daily_ret"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"年化報酬率: 0.360\n",
"年化標準差: 0.259\n",
"年化夏普: 1.3869\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>年/月</th>\n",
" <th>投組日報酬</th>\n",
" <th>投組累積報酬</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>2017-02-03</td>\n",
" <td>0.019334</td>\n",
" <td>1.019334</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2017-02-06</td>\n",
" <td>0.020656</td>\n",
" <td>1.040389</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2017-02-07</td>\n",
" <td>0.000323</td>\n",
" <td>1.040725</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2017-02-08</td>\n",
" <td>0.003820</td>\n",
" <td>1.044701</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2017-02-09</td>\n",
" <td>-0.008778</td>\n",
" <td>1.035531</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>430</th>\n",
" <td>2020-10-26</td>\n",
" <td>-0.008438</td>\n",
" <td>1.769802</td>\n",
" </tr>\n",
" <tr>\n",
" <th>431</th>\n",
" <td>2020-10-27</td>\n",
" <td>0.002837</td>\n",
" <td>1.774823</td>\n",
" </tr>\n",
" <tr>\n",
" <th>432</th>\n",
" <td>2020-10-28</td>\n",
" <td>-0.002829</td>\n",
" <td>1.769802</td>\n",
" </tr>\n",
" <tr>\n",
" <th>433</th>\n",
" <td>2020-10-29</td>\n",
" <td>0.000000</td>\n",
" <td>1.769802</td>\n",
" </tr>\n",
" <tr>\n",
" <th>434</th>\n",
" <td>2020-10-30</td>\n",
" <td>-0.007092</td>\n",
" <td>1.757251</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>435 rows × 3 columns</p>\n",
"</div>"
],
"text/plain": [
" 年/月 投組日報酬 投組累積報酬\n",
"0 2017-02-03 0.019334 1.019334\n",
"1 2017-02-06 0.020656 1.040389\n",
"2 2017-02-07 0.000323 1.040725\n",
"3 2017-02-08 0.003820 1.044701\n",
"4 2017-02-09 -0.008778 1.035531\n",
".. ... ... ...\n",
"430 2020-10-26 -0.008438 1.769802\n",
"431 2020-10-27 0.002837 1.774823\n",
"432 2020-10-28 -0.002829 1.769802\n",
"433 2020-10-29 0.000000 1.769802\n",
"434 2020-10-30 -0.007092 1.757251\n",
"\n",
"[435 rows x 3 columns]"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ret_table, target_stock, daily_ret = condition(financial_data1, 35, 15, 5, 1.5, 15)\n",
"ret_table"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 指標比較"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\terry\\AppData\\Local\\Temp\\ipykernel_7908\\3803379265.py:14: FutureWarning: Using .astype to convert from timezone-aware dtype to timezone-naive dtype is deprecated and will raise in a future version. Use obj.tz_localize(None) or obj.tz_convert('UTC').tz_localize(None) instead\n",
" y9997 = y9997.astype({'年/月':'datetime64[ns]'})\n"
]
}
],
"source": [
"y9997 = tej.get('TWN/APRCD1', \n",
" coid='y9997', \n",
" mdate={'gt': '2017-02-01', 'lt':'2022-10-14'}, \n",
" opts={'columns': ['coid', 'mdate', 'close_adj']},\n",
" chinese_column_name=True, \n",
" paginate=True)\n",
"\n",
"y9997['基準日報酬'] = y9997['收盤價(元)'].pct_change()\n",
"\n",
"y9997['基準累積報酬'] = (y9997['基準日報酬']+1).cumprod()\n",
"\n",
"y9997.rename(columns={'年月日':'年/月'}, inplace=True)\n",
"\n",
"y9997 = y9997.astype({'年/月':'datetime64[ns]'})"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 864x576 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"ret_result1 = pd.merge(y9997, ret_table, how='outer')\n",
"\n",
"ret_result1['投組累積報酬'].fillna(method='ffill', inplace=True)\n",
"\n",
"fig, ax = plt.subplots(figsize=(12, 8))\n",
"\n",
"ax.plot(ret_result1['年/月'], ret_result1['投組累積報酬'], label='投組累積報酬', color='red')\n",
"\n",
"ax.plot(ret_result1['年/月'], ret_result1['基準累積報酬'], label='大盤報酬指數', color='blue')\n",
"\n",
"ax.legend(loc='upper left', fontsize=12)\n",
"\n",
"ax.set_title('費雪策略(PSR1.5倍) vs. Y9997', fontsize=15)\n",
"ax.set_xlabel('時間')\n",
"ax.set_ylabel('累積報酬率')\n",
"\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"年化報酬率: 0.681\n",
"年化標準差: 0.333\n",
"年化夏普: 2.0448\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>年/月</th>\n",
" <th>投組日報酬</th>\n",
" <th>投組累積報酬</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>2017-02-03</td>\n",
" <td>0.019334</td>\n",
" <td>1.019334</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2017-02-06</td>\n",
" <td>0.020656</td>\n",
" <td>1.040389</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2017-02-07</td>\n",
" <td>0.000323</td>\n",
" <td>1.040725</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2017-02-08</td>\n",
" <td>0.003820</td>\n",
" <td>1.044701</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2017-02-09</td>\n",
" <td>-0.008778</td>\n",
" <td>1.035531</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>570</th>\n",
" <td>2020-10-26</td>\n",
" <td>-0.008438</td>\n",
" <td>4.204292</td>\n",
" </tr>\n",
" <tr>\n",
" <th>571</th>\n",
" <td>2020-10-27</td>\n",
" <td>0.002837</td>\n",
" <td>4.216221</td>\n",
" </tr>\n",
" <tr>\n",
" <th>572</th>\n",
" <td>2020-10-28</td>\n",
" <td>-0.002829</td>\n",
" <td>4.204292</td>\n",
" </tr>\n",
" <tr>\n",
" <th>573</th>\n",
" <td>2020-10-29</td>\n",
" <td>0.000000</td>\n",
" <td>4.204292</td>\n",
" </tr>\n",
" <tr>\n",
" <th>574</th>\n",
" <td>2020-10-30</td>\n",
" <td>-0.007092</td>\n",
" <td>4.174477</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>575 rows × 3 columns</p>\n",
"</div>"
],
"text/plain": [
" 年/月 投組日報酬 投組累積報酬\n",
"0 2017-02-03 0.019334 1.019334\n",
"1 2017-02-06 0.020656 1.040389\n",
"2 2017-02-07 0.000323 1.040725\n",
"3 2017-02-08 0.003820 1.044701\n",
"4 2017-02-09 -0.008778 1.035531\n",
".. ... ... ...\n",
"570 2020-10-26 -0.008438 4.204292\n",
"571 2020-10-27 0.002837 4.216221\n",
"572 2020-10-28 -0.002829 4.204292\n",
"573 2020-10-29 0.000000 4.204292\n",
"574 2020-10-30 -0.007092 4.174477\n",
"\n",
"[575 rows x 3 columns]"
]
},
"execution_count": 34,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ret_table2, target_stock2, daily_ret2 = condition(financial_data1, 35, 15, 5, 2, 15)\n",
"\n",
"ret_table2"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 864x576 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"ret_result = pd.merge(y9997, ret_table2, how='outer')\n",
"\n",
"ret_result['投組累積報酬'].fillna(method='ffill', inplace=True)\n",
"\n",
"fig, ax = plt.subplots(figsize=(12, 8))\n",
"\n",
"ax.plot(ret_result['年/月'], ret_result['投組累積報酬'], label='投組累積報酬', color='red')\n",
"\n",
"ax.plot(ret_result['年/月'], ret_result['基準累積報酬'], label='大盤報酬指數', color='blue')\n",
"\n",
"ax.legend(loc='upper left', fontsize=12)\n",
"\n",
"ax.set_title('2倍費雪策略 vs. Y9997', fontsize=15)\n",
"ax.set_xlabel('時間')\n",
"ax.set_ylabel('累積報酬率')\n",
"\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th>公司</th>\n",
" <th>3078</th>\n",
" <th>4908</th>\n",
" <th>8299</th>\n",
" <th>6173</th>\n",
" <th>3118</th>\n",
" <th>6494</th>\n",
" <th>6538</th>\n",
" <th>6577</th>\n",
" </tr>\n",
" <tr>\n",
" <th>年/月</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2019-08-01</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-08-01</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>1.000000</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-08-02</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.973010</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-08-05</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.876420</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-08-06</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.906250</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-10-25</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>1.797221</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-10-28</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>1.818787</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-10-29</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>1.782843</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-10-30</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>1.876298</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2019-11-01</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>1.833165</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>63 rows × 8 columns</p>\n",
"</div>"
],
"text/plain": [
"公司 3078 4908 8299 6173 3118 6494 6538 6577\n",
"年/月 \n",
"2019-08-01 NaN NaN NaN NaN NaN NaN NaN NaN\n",
"2019-08-01 NaN NaN NaN NaN NaN NaN 1.000000 NaN\n",
"2019-08-02 NaN NaN NaN NaN NaN NaN 0.973010 NaN\n",
"2019-08-05 NaN NaN NaN NaN NaN NaN 0.876420 NaN\n",
"2019-08-06 NaN NaN NaN NaN NaN NaN 0.906250 NaN\n",
"... ... ... ... ... ... ... ... ...\n",
"2019-10-25 NaN NaN NaN NaN NaN NaN 1.797221 NaN\n",
"2019-10-28 NaN NaN NaN NaN NaN NaN 1.818787 NaN\n",
"2019-10-29 NaN NaN NaN NaN NaN NaN 1.782843 NaN\n",
"2019-10-30 NaN NaN NaN NaN NaN NaN 1.876298 NaN\n",
"2019-11-01 NaN NaN NaN NaN NaN NaN 1.833165 NaN\n",
"\n",
"[63 rows x 8 columns]"
]
},
"execution_count": 38,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"(pd.concat(daily_ret2)['2019-08-01': '2019-12-01'].pct_change()+1).cumprod()"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Y M \n",
"2017 2 1.081771\n",
" 3 1.088480\n",
" 4 1.090846\n",
" 5 1.148434\n",
" 6 1.264024\n",
" 7 1.219184\n",
" 8 1.418682\n",
" 9 1.533703\n",
"2018 11 1.676307\n",
" 12 1.569611\n",
"2019 1 1.622959\n",
" 2 1.673328\n",
" 3 1.716443\n",
" 4 1.697577\n",
" 5 1.604741\n",
" 6 1.712355\n",
" 7 1.718081\n",
" 12 1.741090\n",
"2020 1 1.654291\n",
" 2 1.636773\n",
" 3 1.501627\n",
" 4 1.847001\n",
" 9 1.739676\n",
" 10 1.757251\n",
"Name: 投組累積報酬, dtype: float64"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ret_table['Y'] = ret_table['年/月'].dt.year\n",
"ret_table['M'] = ret_table['年/月'].dt.month\n",
"ret_table.groupby(['Y', 'M'])['投組累積報酬'].apply(lambda x: x.iloc[-1])"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.9.12 ('base')",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.12"
},
"orig_nbformat": 4,
"vscode": {
"interpreter": {
"hash": "3e06028060a7864164bfee2227bae8dfd39c60041c455d9e6b5a8dd3aec9b09f"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment